Fine-Tuning DARTS for Image Classification

16 Jun 2020  ·  Muhammad Suhaib Tanveer, Muhammad Umar Karim Khan, Chong-Min Kyung ·

Neural Architecture Search (NAS) has gained attraction due to superior classification performance. Differential Architecture Search (DARTS) is a computationally light method. To limit computational resources DARTS makes numerous approximations. These approximations result in inferior performance. We propose to fine-tune DARTS using fixed operations as they are independent of these approximations. Our method offers a good trade-off between the number of parameters and classification accuracy. Our approach improves the top-1 accuracy on Fashion-MNIST, CompCars, and MIO-TCD datasets by 0.56%, 0.50%, and 0.39%, respectively compared to the state-of-the-art approaches. Our approach performs better than DARTS, improving the accuracy by 0.28%, 1.64%, 0.34%, 4.5%, and 3.27% compared to DARTS, on CIFAR-10, CIFAR-100, Fashion-MNIST, CompCars, and MIO-TCD datasets, respectively.

PDF Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Fine-Grained Image Classification CompCars Fine-Tuning DARTS Accuracy 95.9% # 2
Image Classification Fashion-MNIST Fine-Tuning DARTS Percentage error 3.09 # 1
Accuracy 96.91 # 1